CVE-2026-116690 MLflow: Arbitrary file read on unauthenticated MLflow servers via prompt model version source
- Exploited in the wild
- Not known
- Fix
- Available
What is CVE-2026-116690?
CVE-2026-116690 is a high-severity vulnerability in mlflow, published on 27 Oct 2026.
Official description
MLflow is an open source AI engineering platform for agents, large language models, and machine learning models. From 3.10.0 until 3.17.0, the prompt branch in _create_model_version() accepts schemeless relative sources because path validation runs only when a source has a URL scheme, and mlflow.prompt.is_prompt is interpreted by tag presence rather than a true value. A remote caller can store a server-local path as the model version artifact location and use /api/2.0/mlflow/model-versions/get-artifact to read files accessible to the MLflow server process. Default unauthenticated servers require no privileges, while basic-auth deployments are exposed to authenticated users who can create registered models. This issue is fixed in version 3.17.0.
Weakness: CWE-22
Which versions are affected by CVE-2026-116690?
2 open source packages ship an affected version, including mlflow.
| Package | Affected | Fixed in |
|---|---|---|
mlflow PyPI | From 3.10.0 | 3.17.0 |
mlflow PyPI | From 3.10.0 | 3.17.0 |
How do I fix CVE-2026-116690?
Upgrade mlflow to 3.17.0 or later.
Is CVE-2026-116690 being exploited?
There is no known exploitation of CVE-2026-116690 in the wild and no public exploit code.
Sources
Also known as GHSA-m65p-9j7q-6w33.